CXL Consortium releases Compute Express Link 4.0 specification

The CXL Consortium has published the Compute Express Link 4.0 specification to address growing data center workload demands. The release promises doubled bandwidth versus the prior generation and adds new capabilities for coherent memory connectivity.

The CXL Consortium, an industry standard organization that develops and promotes an open coherent interconnection for heterogeneous memory and computing solutions, announced the release of the Compute Express Link 4.0 specification. The specification is presented as a response to increasing demands from emerging workloads placed on todayu2019s data centers, positioning the standard as a foundation for improved memory and device connectivity in heterogeneous architectures.

Derek Rohde, CXL Consortium president and treasurer and principal engineer at NVIDIA, characterized the release as a milestone for advancing coherent memory connectivity. According to the announcement, the CXL 4.0 specification doubles the bandwidth over the previous generation and introduces powerful new features. The consortium framed the update as the result of member companiesu2019 collaboration to produce open standards that enable broader compliance and interoperability across implementations.

The consortium emphasized that the specification is intended to empower data center innovation and help the industry scale for future usage models. By promoting an open, coherent interconnect, the CXL 4.0 release aims to support heterogeneous memory and computing solutions while providing a common framework for compliance and interoperability. The announcement highlights the consortiumu2019s role in coordinating industry efforts to meet evolving workload requirements without detailing specific technical changes beyond the reported bandwidth increase and new feature set.

68

Impact Score

Artificial Intelligence, chips, and robots set the tone at CES 2026

CES 2026 in Las Vegas put Artificial Intelligence at the center of nearly every major announcement, with chipmakers and robotics firms using the show to preview their next wave of platforms and humanoid systems. Nvidia, AMD, Intel, Qualcomm, Google, Samsung, Hyundai, and Boston Dynamics all leaned on Artificial Intelligence to anchor their product strategies.

Inside the UK’s artificial intelligence security institute

The UK’s artificial intelligence security institute has found that popular frontier models can be jailbroken at scale, exposing reliability gaps and security risks for governments and regulated industries that rely on trusted vendors.

Siemens debuts digital twin composer for industrial metaverse deployments

Siemens has introduced digital twin composer, a software tool that builds industrial metaverse environments at scale by merging comprehensive digital twins with real-time physical data, enabling faster virtual decision making. Early deployments with PepsiCo report higher throughput, shorter design cycles and reduced capital expenditure through physics-accurate simulations and artificial intelligence driven optimization.

Contact Us

Got questions? Use the form to contact us.

Contact Form

Clicking next sends a verification code to your email. After verifying, you can enter your message.